Scene-oriented forest fire dynamic risk assessment method

Abstract

The invention relates to the technical field of risk assessment, in particular to a scene-oriented forest fire dynamic risk assessment method, and solves the problems that an existing model depends on existing data, is poor in adaptability to sudden fire or multi-scene conditions, and is difficult to meet real-time and precise assessment requirements. The method comprises the steps of building a forest fire behavior simulation model based on a cellular automaton, building a forest fire scene simulation knowledge graph based on semantic driving, and building a comprehensive forest fire dynamic risk assessment model. According to the method, the risk assessment framework capable of being dynamically updated is constructed through fire behavior simulation, semantic modeling, comprehensive risk assessment model construction and other technologies, real-time monitoring and prediction of fire risks can be achieved through the method, and the disaster prevention and reduction capacity is enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of risk assessment, and particularly to a scenario-oriented dynamic risk assessment method for forest fires. Background Technique

[0002] Forest fires are one of the most common and extremely destructive natural disasters globally. They not only cause great damage to the ecosystem but also pose a serious threat to the safety of human life and property. With the intensification of global warming, changes in land use patterns, and human activities, the frequency and intensity of forest fires show a significant upward trend. How to effectively assess and respond to forest fire risks has become a research focus and technical challenge in the fields of forest management and disaster prevention and control.

[0003] Traditional forest fire risk assessment methods mainly rely on statistical models and physical modeling. Statistical models usually analyze static data, use fire historical records combined with environmental factors such as vegetation types, terrain slopes, and meteorological conditions to quantify risk components such as hazard, exposure, and vulnerability, thereby constructing a quantitative model of fire risk, which can reveal the main driving factors of fire occurrence and provide guidance for long-term risk prediction. However, such models rely on existing data and have poor adaptability to sudden fires or multi-scenario conditions, making it difficult to meet the requirements of real-time and precise assessment. Physical modeling constructs a fire spread model (such as flame propagation path, combustion rate, etc.) by assuming different scenario conditions to simulate the dynamic changes and spread of fires, reflecting the dynamic characteristics of fire spread. Although this method can simulate the real-time changes of fires during the forest fire process, its simulation results are relatively single, and it is insufficient in responding to variable factors under complex environmental conditions.

[0004] Existing risk assessment methods rely on existing data and have poor adaptability to sudden fires or multi-scenario conditions, making it difficult to meet the requirements of real-time and precise assessment under forest fire emergency conditions; therefore, they do not meet the existing needs, and for this reason, we propose a scenario-oriented dynamic risk assessment method for forest fires. Summary of the Invention

[0005] The purpose of the present invention is to provide a scenario-oriented dynamic risk assessment method for forest fires to solve the problem that the existing models rely on existing data, have poor adaptability to sudden fires or multi-scenario conditions, and are difficult to meet the requirements of real-time and precise assessment mentioned in the above background technique.

[0006] To achieve the above purpose, the present invention provides the following technical solution: A scenario-oriented dynamic risk assessment method for forest fires, the method includes:

[0007] Based on the construction of a cellular automaton forest fire behavior simulation model, and based on the requirements of risk assessment, the forest fire prevention and control area is divided into several cells, and the cell states are defined according to different fire behaviors during the forest fire combustion process; the Wang Zhengfei forest fire spread model is used to calculate the forest fire spread rate, and based on the states of adjacent cells and the advancement of the spread time, different cell state transition rules are defined, so as to realize the real-time fire behavior simulation during the forest fire spread process;

[0008] Based on the construction of a semantic-driven knowledge graph for forest fire scenario simulation, guided by the business in the field of forest fires, relevant theories, and the SSN and TIME ontology standards published by W3C, the theoretical framework of the knowledge graph for forest fire scenario simulation is constructed, and the integration and dynamic update of multi-source information in the experimental area and the encapsulation of the forest fire behavior simulation method based on cellular automata are realized by using the semantic modeling method, so as to realize the construction and update of the knowledge graph for forest fire scenario simulation;

[0009] Based on the construction of a comprehensive forest fire dynamic risk assessment model, based on the three risk factors of the likelihood of fire occurrence, the exposure of the disaster-bearing body, and vulnerability, real-time fire environment factors such as the forest fire spread rate, ignition state, and population density are introduced to evaluate these three risk factors, and the forest fire dynamic risk assessment modeling is completed by constructing a comprehensive risk assessment index;

[0010] Based on the scenario simulation-based dynamic risk assessment, the semantic prevention and control simulation scenario is set in the knowledge graph for forest fire scenario simulation, and the above-mentioned cellular automaton forest fire behavior simulation model is called to realize the real-time fire behavior simulation. Finally, based on the forest fire dynamic risk assessment model, real-time risk assessment indicators are introduced to realize the dynamic assessment of the fire risk at different times and prevention and control scenarios;

[0011] Through the simulation of real fire comparison experiments, the executability and prediction accuracy of the described scenario-oriented forest fire dynamic risk assessment method are evaluated.

[0012] Preferably, there are 5 cell states of different fire behaviors in the construction method of the cellular automaton forest fire behavior simulation model, and the fire behavior state characteristics presented by the cells are as follows:

[0013] When the cell state = 0, that is, the cell state S0: the cell is in a fire-separated area, where there is no combustible material in the fire-separated area, and the cell state remains unchanged;

[0014] When the cell state = 1, that is, the cell state S1: the combustible material in the cell is in an unignited state;

[0015] When the cell state = 2, that is, the cell state S2: the combustible material in the cell has just started to burn, and the combustible material is in an internal development state and does not have the ability to ignite the remaining cells outward;

[0016] When the cell state = 3, that is, the cell state S3: the combustible in the cell is in a fully burning state, and the combustible begins to spread outward and has the ability to ignite the remaining cells;

[0017] When the cell state = 4, that is, the cell state S4: the combustible in the cell is in a burned state.

[0018] Preferably, for the conversion rule of the cell state S0 to S1, the non-combustible areas covered by water and roads in the cell are defined as S0, and the combustible areas covered by forests in the cell are defined as S1.

[0019] Preferably, for the conversion rule of the cell state S1 to S2, the unburned S1 state cell is ignited by the influence of the surrounding fully burning S3 state cells. When the cumulative value of the S1 state cell transmission exceeds the threshold, it is ignited. The threshold is 0.65, and the unburned S1 state cell is ignited and converted into the S2 state;

[0020]

[0021] Where is the cumulative ignition transfer value of adjacent cells to the current cell after the next time interval Δt; is the spread rate of cells with the ability to spread in the Moore neighborhood at the current position, that is, cells in the fully burning state S3; Δt is the simulation time step and is set to 1 min, L is the horizontal distance between cells, and i and j are the coordinate indices of the current cell;

[0022] R o : Initial spread speed; K s : Combustible correction coefficient; K w : Wind speed correction coefficient; Terrain correction coefficient.

[0023] Preferably, for the conversion rule of the cell state S2 to S3, the time process of the cell from the internal combustion S2 state to the fully burning S3 state is defined as Δt 1 , and the Δt 1 is used to describe the internal forest fire spread process of the cell. When the combustion duration t cost of the S2 state cell is greater than Δt 1 then the cell state is converted from S2 to S3;

[0024]

[0025] Where R IN is the internal forest fire spread rate of the cell. The cells in S2 are converted into S3 after Δt 1 and have the ability to spread to the neighborhood.

[0026] Preferably, the conversion rule from the cell state S3 to S4 defines the process of the cell transitioning from the fully burning S3 state to the extinguished S4 state as Δt 2 , where the Δt 2 depends on the continuous combustion of the fuel, external fire extinguishing measures, and environmental conditions. The flame combustion duration is related to the time Δt burn required for the complete combustion of the fuel and the fire extinguishing intervention time Δt inter ;

[0027] Δt 2 = Δt burn + Δt inter ;

[0028]

[0029] where W is the fuel load within the cell range and is in kg / m 2 , m is the mass burning rate of the combustible and is in kg / (m 2 ·min).

[0030] Preferably, the forest fire dynamic risk assessment model is based on three risk factors: the likelihood of fire occurrence, the exposure of the disaster-bearing body, and vulnerability, and constructs an index system for forest fire risk assessment. In this index system, the comprehensive risk is defined as the A-level index, the likelihood, exposure, and vulnerability are defined as the B-level risk indexes, and the fire environment factors such as real-time fire behavior, terrain, and vegetation cover are listed as the C-level indexes. During the risk assessment process, by introducing and dynamically updating real-time data such as fire behavior and population distribution in the index system, the dynamic assessment of forest fire risk is achieved;

[0031]

[0032] where R t is the A-level index of the comprehensive risk at time t, H t , E t , V t are the B-level risk indexes of the likelihood, exposure, and vulnerability at time t respectively, BI t and CI t are the B and C-level indexes of the risk at time t respectively, W i is the weight of the C-level index for the B-level index, and n is the number of C-level indexes participating in the B-level index assessment.

[0033] Preferably, among the B-level indexes, the likelihood is calculated from C-level indexes such as the distance from the fire site, combustion state, and combustion rate, and their weight settings are 0.4, 0.3, and 0.3 respectively.

[0034] Preferably, the exposure in the B-level indicators is calculated from the C-level indicators such as population density, distance from buildings, distance from key protected facilities, vegetation coverage rate, GDP, etc., and their weight settings are 0.3, 0.2, 0.25, 0.125, and 0.125 respectively.

[0035] Preferably, the vulnerability of the B-level indicators is calculated from the C-level indicators such as flammable material type, continuous fire duration, etc., and their weight settings are 0.55 and 0.45 respectively.

[0036] Preferably, the stage values of the A-level, B-level, and C-level indicators are evenly divided into five grades from I to V according to the contribution degree of the fire impact. The values of the five grades from I to V are assigned 0.2, 0.4, 0.6, 0.8, and 1 respectively. The higher the grade of the risk indicator, the greater the impact of the fire.

[0037] Preferably, the setting of the fire simulation scenario includes the setting of emergency resources such as the start time of the fire, the end time of the fire, the emergency rescue time, the fire location, the isolation belt, firefighters, fire trucks, and the number of vehicles.

[0038] Preferably, the specific steps of the dynamic risk assessment method based on scenario simulation are as follows: First, integrate multi-source data in the experimental area to complete the dynamic update of the knowledge graph of forest fire scenario simulation, and set the fire simulation scenario based on the knowledge graph; then, call the above-mentioned cellular automata forest fire behavior simulation model according to different fire simulation scenarios to realize the simulation of real-time fire behavior; finally, based on the forest fire dynamic risk assessment model, introduce real-time risk assessment indicators to evaluate the fire risks at different times and prevention and control scenarios.

[0039] Compared with the prior art, the beneficial effects of the present invention are:

[0040] 1. By introducing the concepts of scenario simulation and dynamic data fusion, the present invention breaks through the limitations of traditional statistical models and physical modeling. By combining multi-source data, it can more comprehensively and accurately reflect the fire risk situation, effectively improving the accuracy of fire risk assessment and its applicability in different scenarios. Based on the cellular automata method, it simulates the propagation path and speed of the fire in multi-dimensional space, details the cell states and clarifies their state transition rules, which can deeply reveal the mechanism of fire dynamic changes and help researchers and relevant decision-makers better understand the process and laws of fire development.

[0041] 2. The present invention constructs a risk assessment index system based on the risk of fire occurrence, the exposure of disaster-bearing bodies, and vulnerability. When conducting the assessment, information such as real-time fire behavior and population distribution is introduced to achieve dynamic assessment of fire risk and meet the real-time requirements of emergency management. By setting scenarios such as fire occurrence, environment, and emergency measures through semantic drive, fire spread simulation and risk prediction are carried out based on different prevention and control scenarios, providing comprehensive decision-making support for the fire department and the government, helping to quickly select appropriate prevention and control strategies, and reducing losses. Description of the Drawings

[0042] Figure 1 It is a schematic diagram of the cell state matrix of the present invention;

[0043] Figure 2 It is a flow chart of the cell state conversion of forest fire spread of the present invention;

[0044] Figure 3 It is a flow chart for constructing a knowledge graph of forest fire scenario simulation of the present invention;

[0046] Figure 4 It is a schematic diagram of setting nodes for scenario simulation of the knowledge graph of the present invention;

[0047] Figure 5 It is a schematic diagram of the structure of forest fire risk assessment indicators of the present invention;

[0048] Figure 6 It is a schematic diagram of the research location for scenario setting in an embodiment of the present invention;

[0049] Figure 7 It is a schematic diagram of setting simulation scenarios for three prevention and control measures in an embodiment of the present invention;

[0050] Figure 8 It is a simulation diagram of fire behavior at different times of the 330 fire in Yanjishan in an embodiment of the present invention, (a) Fire behavior simulation diagram at 13:00, (b) Fire spread behavior diagram at 15:00, (c) Fire behavior simulation diagram at 17:00;

[0051] Figure 9 It is a simulation diagram of fire behavior at 17:00 for three prevention and control scenarios in an embodiment of the present invention, (a) Fire behavior simulation diagram of Scenario 1, (b) Fire behavior simulation diagram of Scenario 2, (c) Fire behavior simulation diagram of Scenario 3;

[0052] Figure 10 It is a risk assessment diagram at different times of the 330 fire in Yanjishan in an embodiment of the present invention, (a) Risk assessment diagram at 12:00, (b) Risk assessment diagram at 15:00, (c) Risk assessment diagram at 17:00;

[0053] Figure 11It is the risk assessment diagram at 17 moments for three prevention and control scenarios in the embodiments of the present invention. (a) Risk assessment diagram for Scenario 1, (b) Risk assessment diagram for Scenario 2, (c) Risk assessment diagram for Scenario 3;

[0054] Figure 12 It is the flow chart of the whole present invention. Detailed implementation manners

[0055] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments.

[0056] Please refer to Figures 1 to 5 , an embodiment provided by the present invention: A scenario-oriented dynamic risk assessment method for forest fires, the method includes:

[0057] Based on the construction of a cellular automaton forest fire behavior simulation model, according to the requirements of risk assessment, the forest fire prevention and control area is divided into several cells, and the cell states are defined according to different fire behaviors during the forest fire combustion process; the Wang Zhengfei forest fire spread model is used to calculate the forest fire spread rate, and based on the states of adjacent cells and the advancement of the spread time, different cell state transition rules are defined, so as to realize the real-time fire behavior simulation during the forest fire spread process;

[0058] Based on the construction of a semantic-driven knowledge graph for forest fire scenario simulation, guided by the business and related theories in the field of forest fires and the SSN and TIME ontology standards published by W3C, the theoretical framework of the knowledge graph for forest fire scenario simulation is constructed, and a semantic modeling method is used to realize the integration and dynamic update of multi-source information in the experimental area, as well as the encapsulation of the cellular automaton forest fire behavior simulation method, so as to realize the construction and update of the knowledge graph for forest fire scenario simulation;

[0059] Based on the construction of a comprehensive dynamic risk assessment model for forest fires, based on the three risk factors of the danger of fire occurrence, the exposure of the disaster-bearing body, and vulnerability, real-time fire environment factors such as the forest fire spread rate, ignition state, and population density are introduced to evaluate these three risk factors, and a comprehensive risk assessment index is constructed to complete the dynamic risk assessment modeling of forest fires;

[0060] Based on the dynamic risk assessment of scenario simulation, the semantic prevention and control simulation scenarios are set in the knowledge graph for forest fire scenario simulation, and the above-mentioned cellular automaton forest fire behavior simulation model is called to realize the real-time fire behavior simulation. Finally, based on the dynamic risk assessment model for forest fires, real-time risk assessment indicators are introduced to realize the dynamic assessment of the fire risk at different times and under different prevention and control scenarios;

[0061] By conducting a comparative experiment simulating a real fire, the feasibility and prediction accuracy of the described scenario-oriented dynamic risk assessment method for forest fires are evaluated.

[0062] Please refer to Figure 1 and Figure 2 When the cell state = 0, that is, cell state S0: The cell is in a fire separation area, where there is no combustible material in the fire separation area, and the cell state remains unchanged;

[0063] When the cell state = 1, that is, cell state S1: The combustible material in the cell is in an unignited state;

[0064] When the cell state = 2, that is, cell state S2: The combustible material in the cell has just started to burn, and the combustible material is in an internal development state and does not have the ability to ignite the remaining cells outward;

[0065] When the cell state = 3, that is, cell state S3: The combustible material in the cell is in a fully burning state, and the combustible material begins to spread outward and has the ability to ignite the remaining cells;

[0066] When the cell state = 4, that is, cell state S4: The combustible material in the cell is in a burned state.

[0067] Among them, the conversion rule from cell state S0 to S1 defines the fire separation area covered by water and roads of the cell as S0, and the non-fire separation area covered by forest of the cell as S1; the conversion rule from cell state S1 to S2 is that the unburned S1 state cell is ignited by the influence of the fully burning S3 state cell around it, and when the cumulative value of the S1 state cell transmission exceeds the threshold, it is ignited. The threshold is 0.65, and the unburned S1 state cell is ignited and converted into the S2 state;

[0068]

[0069] Among them is the cumulative ignition transmission value of adjacent cells to the current cell after the next time interval Δt; is the spread rate of cells with the ability to spread in the Moore neighborhood of the current position, that is, cells in the fully burning state S3; Δt is the simulation time step and is set to 1 min, L is the horizontal distance between cells, and i and j are the coordinate indices of the current cell;

[0070] R o : Initial spread speed; K s : Combustible correction coefficient; K w : Wind speed correction coefficient; Terrain correction coefficient.

[0071] The conversion rule from cell state S2 to S3 defines the time process from the internal combustion S2 state of the cell to the fully burning S3 state as Δt 1, Δt 1 used to describe the internal forest fire spread process of the cell. When the burning duration t of the cell in state S2 cost is greater than Δt 1 then the cell state is converted from S2 to S3;

[0072]

[0073] where R IN is the internal forest fire spread rate of the cell. The cell in S2 is converted to S3 after Δt 1 and has the ability to spread to the neighborhood.

[0074] Cell state S3 to S4 conversion rule. The process of converting the cell from the fully burning S3 state to the extinguished S4 state is defined as Δt 2 , Δt 2 depends on the continuous burning of the fuel, external fire extinguishing measures and environmental conditions. The flame burning duration is related to the time Δt burn required for the complete combustion of the fuel and the fire extinguishing intervention time Δt inter ;

[0075] Δt 2 = Δt burn + Δt inter ;

[0076]

[0077] where W is the fuel load within the cell range and the unit is kg / m 2 , m is the mass burning rate of the combustible and the unit is kg / (m 2 ·min).

[0078] Please refer to Figure 3 , use Protégé software to complete the ontology knowledge modeling including the forest fire element monitoring module and forest fire knowledge elements, adopt Neo4j graph database for the integration and formal display of ontology associations, and then realize the construction and shaping of the forest fire scenario simulation knowledge graph. Based on multi-source data such as remote sensing, meteorology, and ground surveys, use GIS tools and Java coding to extract the fire environment scenario characteristics such as weather conditions, terrain features, and vegetation types with geographical cells as units, and realize the dynamic update of the surface elements and knowledge element entities in the forest fire scenario simulation knowledge graph through Java interfaces to complete the construction of the forest fire environment scenario.

[0079] Please refer to Figure 4 to complete the setting of the forest fire scenario simulation, set the initial conditions for the occurrence of the fire, determine the fire source location and ignition time, and at the same time set the types of prevention and control measures such as physical isolation measures and the input of fire fighting forces, as well as the fire discovery time and the time when emergency rescue and fire fighting start to take effect at the scene. Among them, only the isolation belt and firefighters, fire trucks, and water trucks are set for the emergency prevention and control measures, and the isolation belt needs to be preset in advance according to the development state of the fire.

[0080] Please refer to Figure 5 , the forest fire dynamic risk assessment model is based on three risk factors: the risk of fire occurrence, the exposure of the disaster-bearing body, and vulnerability, and constructs an index system for forest fire risk assessment. In this index system, the comprehensive risk is defined as the A-level index, the risk, exposure, and vulnerability are defined as the B-level risk indicators, while the fire environment factors such as real-time fire behavior, terrain, and vegetation coverage are listed as the C-level indicators. In the process of risk assessment, by introducing and dynamically updating real-time data such as fire behavior and population distribution in the index system, the dynamic assessment of forest fire risk is realized;

[0081]

[0082]

[0083] Among them, R t is the A-level index of the comprehensive risk at time t, H t , E t , V t are the B-level risk indicators of risk, exposure, and vulnerability at time t respectively, BI t and CI t are the B and C-level indicators of risk at time t respectively, W i is the weight of the C-level indicator for the B-level indicator, and n is the number of C-level indicators participating in the B-level indicator assessment;

[0084] Among the B-level indicators, the risk is calculated from C-level indicators such as the distance from the fire site, combustion state, and combustion rate, and their weight settings are 0.4, 0.3, and 0.3 respectively. Among the B-level indicators, the exposure is calculated from C-level indicators such as population density, distance from buildings, distance from key protection facilities, vegetation coverage rate, and GDP, and their weight settings are 0.3, 0.2, 0.25, 0.125, and 0.125 respectively. The vulnerability of the B-level indicator is calculated from C-level indicators such as the type of flammable materials and the continuous ignition time, and their weight settings are 0.55 and 0.45 respectively;

[0085] The stage values of the A-level, B-level, and C-level indicators are evenly divided into five grades from I to V according to the contribution degree of the fire impact. The values assigned to the five grades from I to V are 0.2, 0.4, 0.6, 0.8, and 1 respectively. The higher the grade of the risk indicator, the greater the impact of the fire.

[0086] The stage values of the A-level, B-level, and C-level indicators are evenly divided into five grades from I to V according to the contribution degree to the fire impact. The values assigned to the five grades from I to V are 0.2, 0.4, 0.6, 0.8, and 1 respectively. The higher the grade of the risk indicator, the greater the impact of the fire.

[0087] The C-level indicators for evaluating the hazard index and their grade divisions and value assignments are shown in Table 1 below:

[0088] Table 1

[0089]

[0090]

[0091] The C-level indicators for evaluating the exposure index and their grade divisions and value assignments are shown in Table 2:

[0092] Table 2

[0093]

[0094] The C-level indicators for evaluating the vulnerability index and their grade divisions and value assignments are shown in Table 3 below:

[0095] Table 3

[0096] Flammable material type Fire duration Index level Assignment Others 0-10 I 0.2 Cultivated land 10-30 II 0.4 Shrub forest 30-60 III 0.6 Broad-leaved forest 60-90 IV 0.8 Coniferous forest, building >90 V 1

[0097] Since the contribution degree of each C-level indicator to the B-level risk indicators such as hazard, exposure, and vulnerability is different, it is necessary to set weights for each C-level indicator. The weights of the C-level indicators are set as shown in Table 4 below:

[0098] Table 4

[0099]

[0100]

[0101] Based on the evaluation of the importance of each indicator, after scoring and weighted averaging, the weights of each indicator are obtained. Then, during the dynamic risk assessment process, personnel dynamically adjust the weights through real-time data feedback to make the model more accurate.

[0102] The steps of the dynamic risk assessment method based on scenario simulation are as follows: First, integrate multi-source data in the experimental area to complete the dynamic update of the knowledge graph for forest fire scenario simulation, and set fire simulation scenarios based on the knowledge graph; then, call the above-mentioned cellular automata forest fire behavior simulation model according to different fire simulation scenarios to achieve real-time fire behavior simulation; finally, based on the forest fire dynamic risk assessment model, introduce real-time risk assessment indicators to evaluate the fire risks at different times and prevention and control scenarios.

[0103] Example:

[0104] Taking the "3.30" fire in Yanjishan Scenic Area on March 30, 2019 as an example, the Yanjishan Scenic Area and its surrounding areas are selected as the study area (as Figure 6 shown). By realizing the semantics-driven fire spread simulation to construct the in-real-situation fire scenario during the fire, comparing the real-time fire spread simulation with the relevant reports during the fire occurrence, the method proposed in this study is verified for its effectiveness in predicting the WUI fire spread, providing a scientific basis for fire management and emergency decision-making;

[0105] The simulation scenarios at 13:00, 15:00, and 17:00 after the fire without emergency measures and under three emergency prevention and control measures are simulated (the three prevention and control scenarios can be seen in Table 5, and the setting of the knowledge graph prevention and control scenario can be seen in Figure 7 ). The simulation results are as Figure 8 and Figure 9 shown. At 13:00, the fire spread to Pinggu District, and at about 17:00, the fire was approaching the Millennium Monument Bixia Yuanjun Temple. The simulation results are basically consistent with the fire spread trend in the news reports. It can be considered that the model can relatively accurately simulate the diffusion process of the study area.

[0106] Table 5

[0107]

[0108] By implementing the fire behavior data and other data at different simulated times, the fire risks at 12:00 before the fire, 15:00 and 17:00 after the fire are evaluated (as Figure 10 shown), and the fire risks at 17:00 under the above three set prevention and control scenarios, as Figure 11 shown. In this article, the fire risk is between 0 and 1, and the higher the value, the greater the impact of the fire. From Figure 10 and Figure 11 , it can be seen that through the evaluation of the dynamic risks under different prevention and control scenarios, it is clear that reasonable and effective prevention and control measures can significantly reduce the WUI fire risk. In actual emergency decision-making, the dynamic risk assessment system in this article can be used to evaluate the regional fire risks under different prevention and control measure scenarios and select the most appropriate prevention and control strategy to minimize the harm caused by the fire.

[0109] It is obvious to those skilled in the art that the present invention is not limited to the details of the above-described exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be embraced within the present invention. Any reference signs in the claims should not be construed as limiting the claims involved.

Claims

1. A scenario-oriented forest fire dynamic risk assessment method, characterized in that: The method comprises: Based on the construction of the cellular automaton forest fire behavior simulation model, the forest fire prevention and control area is divided into several cells based on the needs of risk assessment, and the cell states are defined according to the different fire behaviors during the forest fire burning process; the forest fire spread rate is calculated using Wang Zhengfei's forest fire spread model, and based on the states of adjacent cells and the advancement of the spread time, the state transition rules of different cells are defined, thereby realizing real-time fire behavior simulation during the forest fire spread process; The construction of the forest fire scenario simulation knowledge graph based on semantics is guided by the forest fire business and related theories as well as the SSN and TIME ontology standards published by W3C. The theoretical framework of the forest fire scenario simulation knowledge graph is completed. The semantic modeling method is used to realize the integration and dynamic update of multi-source information in the experimental area, and the encapsulation of the forest fire behavior simulation method based on cellular automation, so as to realize the construction and update of the forest fire scenario simulation knowledge graph. The construction of a comprehensive forest fire dynamic risk assessment model is based on the three risk factors of fire hazard, exposure and vulnerability of the disaster-bearing body. Real-time fire environmental factors such as fire spread rate, fire status and personnel density are introduced to assess the three risk factors. The dynamic risk assessment modeling of forest fire is completed by constructing comprehensive risk assessment indicators. Based on the dynamic risk assessment of scenario simulation, the semantic prevention and control simulation scenario is set in the forest fire scenario simulation knowledge graph, and the above-mentioned cellular automaton forest fire behavior simulation model is called to realize the simulation of real-time fire behavior. Finally, based on the dynamic risk assessment model of forest fire, real-time risk assessment indicators are introduced to realize the dynamic assessment of fire risks at different times and prevention and control scenarios. By simulating real fire comparative experiments, the feasibility and prediction accuracy of the scenario-oriented forest fire dynamic risk assessment method were evaluated.

2. A scenario-oriented forest fire dynamic risk assessment method according to claim 1, characterized in that: There are five different fire behavior cellular states in the method for constructing the cellular automaton forest fire behavior simulation model. The fire behavior state characteristics presented by the cells are as follows: When the cell state = 0, that is, the cell state S0: the cell is in the fire isolation area, where there is no combustible material in the fire isolation area, and the cell state remains unchanged; When the cell state = 1, that is, the cell state S1: the combustible material in the cell is in a state where it has not been ignited; When the cell state = 2, that is, the cell state S2: the combustibles in the cell have just started to burn, and the combustibles are in the internal development state and have no ability to ignite other cells outward; When the cell state = 3, that is, the cell state S3: the combustibles in the cell are in a state of complete combustion, and the combustibles begin to diffuse outward and have the ability to ignite the remaining cells; When the cell state = 4, that is, the cell state S4: the combustible material in the cell is in a burned state.

3. A scenario-oriented forest fire dynamic risk assessment method according to claim 2, characterized in that: The rule for converting the cell state S0 to S1 defines the fire-proof area where the cells cover water areas and roads as S0, and defines the non-fire-proof area where the cells cover forests as S1.

4. A scenario-oriented forest fire dynamic risk assessment method according to claim 3, characterized in that: The cell state S1 to S2 conversion rule, the combustible unburned S1 state cell is ignited by the surrounding completely burned S3 state cells, and is ignited when the S1 state cell transfer cumulative value exceeds the threshold, the threshold is 0.65, the unburned S1 state cell is ignited and converted to S2 state; in is the cumulative value of the ignition transfer from the adjacent cells to the current cell after the next time interval Δt; is the spreading rate of cells with spreading ability, i.e., cells in the complete combustion state S3, in the Moore neighborhood of the current position; Δt is the simulation time step and is set to 1min, L is the horizontal distance between cells, and i and j are the coordinate indexes of the current cell; R o : initial spreading speed; K s : Combustible correction factor; K w : Wind speed correction coefficient; Terrain correction factor.

5. A scenario-oriented forest fire dynamic risk assessment method according to claim 4, characterized in that: The cell state S2 to S3 transition rule defines the time process of the cell from the internal combustion S2 state to the complete combustion S3 state as Δt1, and the Δt1 is used to describe the internal forest fire spread process of the cell. When the combustion duration of the S2 state cell is t cost When it is greater than Δt1, the cell state changes from S2 to S3; Where R IN is the internal forest fire spread rate of the cell. The cells in S2 are transformed into S3 after Δt1 and have the ability to spread to the neighborhood.

6. A scenario-oriented forest fire dynamic risk assessment method according to claim 5, characterized in that: The cell state S3 to S4 transition rule defines the process of the cell from the complete combustion S3 state to the extinguished S4 state as Δt2, which depends on the continuous combustion of the fuel, external fire extinguishing measures and environmental conditions. The flame combustion duration is related to the time Δt required for the complete combustion of the fuel. burn and fire extinguishing intervention time Δt inter Related; Δt2=Δt burn +Δt inter ; Where W is the fuel load within the cell and is expressed in kg / m 2 , m is the mass combustion rate of the combustible material, and the unit is kg / (m 2 ·min).

7. A scenario-oriented forest fire dynamic risk assessment method according to claim 6, characterized in that: The dynamic risk assessment model for forest fires is based on the three risk factors of fire hazard, exposure and vulnerability of the disaster-bearing body, and constructs an index system for forest fire risk assessment. In this index system, comprehensive risk is defined as a Class A index, hazard, exposure and vulnerability are defined as Class B risk indicators, and real-time fire behavior, terrain, vegetation coverage and other fire environment factors are listed as Class C indicators. In the risk assessment process, real-time data such as fire behavior and population distribution are introduced and dynamically updated in the index system to achieve dynamic assessment of forest fire risk; Where R t is the A-level index of comprehensive risk at time t, H t 、E t 、V t are the risk B indicators of hazard, exposure and vulnerability at time t, BI t and CI t are the B and C level indicators of risk at time t, respectively, i is the weight of the C-level indicator to the B-level indicator, and n is the number of C-level indicators participating in the evaluation of the B-level indicators; The danger level in the B-level index is calculated by the C-level indexes such as the distance from the fire scene, the burning state, and the burning rate, and the weights thereof are set to 0.4, 0.3, and 0.3 respectively; The exposure in the B-level index is calculated by the C-level indexes such as population density, distance to buildings, distance to key protection facilities, vegetation coverage, and GDP, and the weights are set to 0.3, 0.2, 0.25, 0.125, and 0.125 respectively; The B-level indicator vulnerability is calculated based on the C-level indicators such as the type of flammable material and the duration of ignition, and the weights thereof are set to 0.55 and 0.45 respectively.

8. A scenario-oriented forest fire dynamic risk assessment method according to claim 7, characterized in that: The stage values ​​of the A-level indicators, B-level indicators and C-level indicators are divided into five levels from I to V according to the degree of contribution to the impact of fire. The five levels from I to V are assigned values ​​of 0.2, 0.4, 0.6, 0.8 and 1 respectively. The higher the level of the risk indicator, the greater the impact of the fire.

9. A scenario-oriented forest fire dynamic risk assessment method according to claim 8, characterized in that: The fire simulation scenario setting includes the setting of emergency resources such as the fire start time, fire end time, emergency rescue time, fire point, isolation belt, firefighters, fire trucks, and number of trucks.

10. A scenario-oriented forest fire dynamic risk assessment method according to claim 9, characterized in that: The specific steps of the dynamic risk assessment method based on scenario simulation are as follows: first, integrate multi-source data of the experimental area to complete the dynamic update of the forest fire scenario simulation knowledge graph, and set the fire simulation scenario based on the knowledge graph; then, according to different fire simulation scenarios, call the above-mentioned cellular automaton forest fire behavior simulation model to realize real-time fire behavior simulation; finally, based on the forest fire dynamic risk assessment model, introduce real-time risk assessment indicators to evaluate the fire risk at different times and prevention and control scenarios.

Images